Discrimination between singing and speech in real-world audio.
Brian ThompsonPublished in: SLT (2014)
Keyphrases
- audio features
- real world
- audio visual
- audio stream
- acoustic features
- broadcast news
- text to speech
- audio signals
- music information retrieval
- emotion recognition
- speaker identification
- data sets
- speech processing
- multimedia
- synthetic data
- speech recognition
- speech signal
- visual features
- audio video
- visual information
- multi stream
- visual speech
- low level
- acoustic signals
- linear predictive coding
- digital audio
- audio recordings
- cepstral features
- speaker recognition
- audio signal
- language acquisition
- automatic speech recognition
- signal processing
- feature set
- natural language
- case study
- neural network
- speech synthesis
- information retrieval systems
- recognition engine
- prosodic features
- wide range
- spoken documents
- data mining